Views
No views yet
instructlab/merlinite-7b-lab and trained using Low-Rank Adaptation (LoRA) to enhance instruction-following capabilities, particularly for AWS User Group and cloud computing-related questions.instructlab/merlinite-7b-lab1from transformers import AutoModelForCausalLM, AutoTokenizer
2from peft import PeftModel
3
4# Load base model
5base_model = AutoModelForCausalLM.from_pretrained("instructlab/merlinite-7b-lab", torch_dtype="auto")
6
7# Load LoRA adapter
8model = PeftModel.from_pretrained(base_model, "mzzavaa/otto_model_demo_1")
9
10# Load tokenizer
11tokenizer = AutoTokenizer.from_pretrained("instructlab/merlinite-7b-lab")
12
13# Test inference
14input_text = "What is the typical time and venue for AWS User Group Vienna meetups?"
15inputs = tokenizer(input_text, return_tensors="pt")
16output = model.generate(**inputs)
17print(tokenizer.decode(output[0], skip_special_tokens=True))ilab data generateinstructlab/merlinite-7b-labLoRA (Low-Rank Adaptation)bf16 for efficiencyAdamW2e-5| Metric | Value |
|---|---|
| Accuracy | [To be tested] |
| Perplexity | [To be tested] |
| F1 Score | [To be tested] |
1@misc{mzzavaa2024otto,
2 title={OTTO Model Demo 1 - LoRA Adapter for AWS},
3 author={mzzavaa},
4 year={2025},
5 howpublished={\url{https://huggingface.co/mzzavaa/otto_model_demo_1}}
6}